Intelligent Decision-making Method for Coating Production Line and Coating Production Line

Abstract

The present invention provides an intelligent decision-making method for a painting production line and a painting production line. The intelligent decision-making method for the painting production line includes: determining the usage life period of the painting components on the painting production line; screening target painting records according to the usage life period, and training an intelligent decision-making model based on the target painting records; performing multi-objective optimization based on the intelligent decision-making model to obtain an intelligent decision-making result; and performing corresponding control of the painting production line based on the intelligent decision-making result. For the intelligent decision-making method for the painting production line and the painting production line of the present invention, the usage duration of the painting components is introduced to determine the usage life period of the painting components, and a suitable intelligent decision-making model is matched according to the usage life period; multi-objective optimization is performed according to the intelligent decision-making model to obtain an intelligent decision-making result, and corresponding control of the painting production line is performed based on the intelligent decision-making result, so that the control of the painting production line is more intelligent and suitable.

Description

Technical Field

[0001] The present invention relates to the technical field of painting, and particularly to an intelligent decision-making method for a painting production line and a painting production line. Background Art

[0002] A painting production line is an industrial production facility dedicated to painting the surfaces of various products (such as automobiles, furniture, mechanical equipment, etc.). The purpose of the painting production line is to evenly apply paint to the product surface through a series of continuous process steps and ensure the drying and curing of the coating to meet the requirements of protection, decoration, or special functions (such as insulation, anti-corrosion, etc.). The design and configuration of the painting production line vary according to factors such as product type, painting requirements, production volume, and technical level. Modern painting production lines increasingly tend to be automated and intelligent to improve efficiency, reduce costs, and ensure quality.

[0003] The invention patent with the application number: CN202110150246.7 discloses a robot vision integration system for a painting production line, including a visual image acquisition module, a communication module, a supplementary lighting module, a visual image processing module, a visual image positioning module, a visual image measurement module, a visual tracking simulation module, a database, a control module, and a spraying route map output module; by setting the visual tracking simulation module, the database, and the control module, a spraying route simulation map can be formed according to the visual information of the actual workpiece, and the spraying route simulation map is matched and searched with the existing spraying route reference teaching map in the database to obtain an optimal target spraying route map as the robot vision map in the painting production line, which can effectively reduce the influence of the visual deviation of the robot on the spraying quality, thereby ensuring that the robot can obtain high spraying quality.

[0004] However, during the spraying operation of the painting production line, in addition to optimizing the spraying accuracy, multiple production line goals (such as spraying cost, spraying efficiency, etc.) need to be considered. The above-mentioned existing technology only considers the optimization of one production line production goal, and is not intelligent enough and has low applicability when facing more diverse optimization requirements.

[0005] In view of this, there is an urgent need for an intelligent decision-making method for a painting production line and a painting production line to at least solve the above deficiencies. Summary of the Invention

[0006] One of the purposes of the present invention is to provide an intelligent decision-making method for a painting production line and a painting production line, introduce the service life of the painting components, determine the service life period of the painting components, and match a suitable intelligent decision-making model according to the service life period; perform multi-objective optimization according to the intelligent decision-making model to obtain an intelligent decision-making result, and perform corresponding control of the painting production line based on the intelligent decision-making result, so that the control of the painting production line is more intelligent and suitable.

[0007] The intelligent decision-making method for a painting production line provided by an embodiment of the present invention includes:

[0008] Determine the usage life period of the painting components on the painting production line;

[0009] According to the usage life period, screen the target painting records, and train an intelligent decision-making model based on the target painting records;

[0010] Perform multi-objective optimization based on the intelligent decision-making model to obtain an intelligent decision-making result;

[0011] Perform corresponding control of the painting production line based on the intelligent decision-making result.

[0012] Preferably, determining the usage life period of the painting components on the painting production line includes:

[0013] Obtain the equipment maintenance records of the painting production line;

[0014] According to the equipment maintenance records, obtain the usage duration of the painting components;

[0015] Obtain the consumable types of the painting components;

[0016] According to the consumable types, determine the life cycle determination rules of the painting components;

[0017] Determine the usage life period according to the life cycle determination rules and the usage duration.

[0018] Preferably, according to the usage life period, screening the target painting records and training the intelligent decision-making model based on the target painting records includes:

[0019] Obtain the preselected painting records; the preselected painting records include: reference painting records obtained from big data and historical painting records on the painting production line;

[0020] Analyze the preselected painting records to determine the reference usage life period of the reference painting components;

[0021] Determine the target painting records according to the usage life period and the reference usage life period;

[0022] Analyze the target painting records to obtain a training data set; the training data set includes: a training painting parameter set and a training painting quality parameter set that correspond one by one;

[0023] Based on a preset CNN model, train the intelligent decision-making model according to the training data set.

[0024] Preferably, performing multi-objective optimization based on the intelligent decision-making model to obtain an intelligent decision-making result includes:

[0025] Based on the intelligent decision-making model, use the optimization algorithm for multi-objective optimization to obtain the intelligent decision-making result.

[0026] Preferably, based on the intelligent decision-making model, use the optimization algorithm for multi-objective optimization to obtain the intelligent decision-making result, including:

[0027] Obtain the optimization objectives;

[0028] Based on the intelligent decision-making model, according to the optimization algorithm and the optimization objectives, determine the Pareto optimal solution set;

[0029] Visualize the Pareto optimal solution set as the Pareto front and present it to the production line managers, and obtain the intelligent decision-making result according to the target solution selected by the production line managers.

[0030] Preferably, visualizing the Pareto optimal solution set as the Pareto front and presenting it to the production line managers includes:

[0031] Show the optimization objectives to the production line managers and obtain the optimization weights of the optimization objectives set by the production line managers;

[0032] According to the Pareto optimal solution set, determine the optimization quality parameter set corresponding to each Pareto optimal solution;

[0033] Obtain the initial quality parameter set;

[0034] According to the optimization quality parameter set and the initial quality parameter set, determine the quality improvement parameter set;

[0035] According to the quality improvement parameter set and the optimization weights, determine the optimization value;

[0036] According to the corresponding relationship between the optimization value and the Pareto optimal solution, mark the optimization value correspondingly in the Pareto front.

[0037] Preferably, based on the intelligent decision-making result, perform corresponding control on the painting production line, including:

[0038] Obtain the current painting parameter set;

[0039] According to the current painting parameter set, the intelligent decision-making result and the preset control factor library, determine the control factors of the painting production line;

[0040] Based on the control factors, control the painting production line to carry out production.

[0041] Preferably, based on the intelligent decision-making result, performing corresponding control on the painting production line further includes:

[0042] Obtain the first painting process of the first painting target on the painting production line;

[0043] Obtain the switching moment of the intelligent decision-making result;

[0044] Determine a second painting target according to the switching moment and the first painting process;

[0045] Extract image features of the second painting target to obtain image features;

[0046] Determine the painting effect of the second painting target according to the image features;

[0047] If the painting effect is less than or equal to a preset painting effect threshold, readjust the second painting target.

[0048] Preferably, extracting image features of the second painting target to obtain image features includes:

[0049] Determine an intervention process point according to the switching moment and the first painting process;

[0050] Determine the intervention painting position of the second painting target according to the intervention process point;

[0051] Obtain the target section corresponding to the intervention painting position for the second painting target;

[0052] Obtain the projection vector of the target section; the straight line where the projection vector is located is perpendicular to the target section, and the vector direction of the projection vector is from the intervention painting position and towards outside the second painting target;

[0053] Obtain the shooting vector of the shooting device;

[0054] If it satisfies that for any projection vector, there is a vector angle between the shooting vector and the projection vector greater than or equal to a preset angle threshold, and when there is a vector angle between the shooting vector and the projection vector greater than or equal to the preset angle threshold, the corresponding shooting devices are all triggered to shoot, then obtain image features according to the corresponding shooting images.

[0055] Preferably, determining the painting effect of the second painting target according to the image features includes:

[0056] Obtain the target attribute of the second painting target;

[0057] Obtain the painting evaluation record according to the target attribute;

[0058] Obtain the painting stage of the second painting target;

[0059] Determine the target evaluation record item in the painting evaluation record according to the painting stage;

[0060] Train an evaluation model according to the target evaluation record item;

[0061] Determine the evaluation prompt of the evaluation model according to the intervention painting position;

[0062] After the evaluation prompt is completed, input the image features into the evaluation model to determine the painting effect of the second painting target.

[0063] The painting production line provided by the embodiment of the present invention applies the steps of intelligent decision-making of the painting production line as described above.

[0064] The beneficial effects of the present invention are as follows:

[0065] The present invention introduces the service life of the painting components, determines the service life period of the painting components, matches a suitable intelligent decision-making model according to the service life period; performs multi-objective optimization according to the intelligent decision-making model to obtain the intelligent decision result, and performs corresponding control of the painting production line based on the intelligent decision result, and the control of the painting production line is more intelligent and suitable.

[0066] Other features and advantages of the present invention will be described in the following specification, and, in part, will be obvious from the specification, or will be understood by implementing the present invention. The objectives and other advantages of the present invention can be achieved and obtained through the structure specifically pointed out in this application document.

[0067] Next, through the drawings and embodiments, the technical solutions of the present invention will be further described in detail. Description of the Drawings

[0068] The drawings are used to provide a further understanding of the present invention, and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation to the present invention. In the drawings:

[0069] Figure 1 It is a schematic diagram of the intelligent decision-making method for the painting production line in the embodiment of the present invention. Detailed Embodiments

[0070] The following describes the preferred embodiments of the present invention with reference to the drawings. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present invention and are not used to limit the present invention.

[0071] The embodiment of the present invention provides an intelligent decision-making method for a painting production line, as Figure 1 shown, including:

[0072] Step 1: Determine the service life period of the painting components on the painting production line; wherein, the painting components are: the components performing painting operations on the painting production line, such as: brushes, spray gun nozzles, and filter meshes on the painting line; the service life is: the time length of the painting components being put into use on the painting production line, obtained according to the equipment maintenance records; the service life period is: different periods within the service cycle of the painting components, such as: initial stage, middle stage, and late stage;

[0073] Step 2: According to the service life period, the target painting records are screened, and the intelligent decision model is trained according to the target painting records; wherein, the target painting records are: the records of the historical painting operations of the reference painting parts that are at the same stage of the service life period, which are obtained through big data or local historical records; when training the intelligent decision model according to the target painting records, the target painting records are parsed to obtain the one-to-one corresponding training painting parameter set and training painting quality parameter set, the training painting parameter set is: various adjustable parameters involved in the painting process, such as: paint ratio, spraying pressure, drying temperature, etc.; the training painting quality parameter set is: indicators for evaluating the quality of painted products, such as: coating film thickness, coating film uniformity, etc.; introduce the CNN model, and train the intelligent decision model according to the training data set (the AI ​​model that determines the coating parameter set according to the required coating quality parameters);

[0074] Step 3: Perform multi-objective optimization based on the intelligent decision model to obtain intelligent decision results; wherein the intelligent decision results are: coating parameters used by coated parts that are suitable for the service life period. When performing multi-objective optimization, perform multi-objective optimization based on the multi-objective optimization algorithm and the intelligent decision model to obtain intelligent decision results;

[0075] Step 4: Control the coating production line accordingly based on the intelligent decision-making results. Among them, the corresponding control of the coating production line based on the intelligent decision-making results includes: obtaining the current coating parameter set; determining the control factor of the coating production line according to the current coating parameter set, the intelligent decision result and the preset control factor library; controlling the coating production line to produce based on the control factor; the control factor library includes the parameter differences and corresponding pre-selected control factors of multiple optimized coating parameter sets and the current coating parameter set, and the pre-selected control factor corresponding to the parameter difference that meets the situation is determined according to the current coating parameter set and the intelligent decision result and used as the control factor; the control factor is the control vector of the coating production line.

[0076] The working principle and beneficial effects of the above technical solution are:

[0077] The present invention introduces the usage time of the painted parts, determines the service life period of the painted parts, and matches a suitable intelligent decision-making model according to the service life period; performs multi-objective optimization according to the intelligent decision-making model to obtain the intelligent decision result, and performs corresponding control of the coating production line based on the intelligent decision result, so that the control of the coating production line is more intelligent and appropriate.

[0078] In one embodiment, according to the service life period, the target painting records are screened, and the intelligent decision model is trained according to the target painting records, including:

[0079] Obtain preselected painting records; the preselected painting records include: reference painting records obtained from big data and historical painting records on the painting production line; among them, the reference painting records are: painting records on the painting production line obtained from big data; the historical painting records are: painting records on the local painting production line;

[0080] Analyze the preselected painting records to determine the reference service life period of the reference painting components; among them, the reference service life period is: the service life stage of each reference painting component in the preselected painting records;

[0081] Determine the target painting records according to the service life period and the reference service life period; among them, the target painting records are: the preselected painting records in which the service life period of each painting component corresponds exactly to the reference service life period of the reference painting component;

[0082] Analyze the target painting records to obtain a training data set; the training data set includes: a one-to-one corresponding training painting parameter set and a training painting quality parameter set;

[0083] Based on a preset CNN model, train the intelligent decision-making model according to the training data set.

[0084] The working principle and beneficial effects of the above technical solution are as follows:

[0085] The present invention obtains preselected painting records in two ways for subsequent training of the intelligent decision-making model, improving the comprehensiveness of the training; matching the service life periods of the painting components and the reference painting components reduces the training error caused by component differences and improves the suitability of the training.

[0086] In one embodiment, multi-objective optimization is performed based on the intelligent decision-making model to obtain an intelligent decision result, including:

[0087] Based on the intelligent decision-making model, use an optimization algorithm to perform multi-objective optimization to obtain an intelligent decision result.

[0088] The working principle and beneficial effects of the above technical solution are as follows:

[0089] The present invention performs multi-objective optimization based on the intelligent decision-making model and uses an optimization algorithm to achieve the dynamic balance of multiple optimization objectives during the generation process, improving the balance of the optimization and contributing to the intelligent and reasonable decision-making of the painting production line.

[0090] In one embodiment, multi-objective optimization is performed based on the intelligent decision-making model, using an optimization algorithm to obtain an intelligent decision result, including:

[0091] Obtain optimization objectives; among them, the optimization objectives are: the preset quality standards for each quality index, such as: how much the painting thickness reaches, how much the painting uniformity reaches, etc., which are pre-entered by the staff;

[0092] Based on the intelligent decision-making model, according to the optimization algorithm and optimization objectives, determine the Pareto optimal solution set; wherein, the Pareto optimal solution set is such that: among a set of solutions, none of the solutions can be improved by other solutions in at least one objective without getting worse in other objectives when there is no other solution that can improve simultaneously in all objectives.

[0093] Visualize the Pareto optimal solution set as the Pareto front and present it to the production line managers, and obtain the intelligent decision-making result according to the target solution selected by the production line managers. Among them, the Pareto front is: the graphical representation of the Pareto optimal solution in the objective space.

[0094] The working principle and beneficial effects of the above technical solution are as follows:

[0095] The present invention determines the Pareto optimal solution set according to the optimization objectives, visualizes the complex Pareto optimal solution set as the Pareto front, and the production line managers can intuitively see the performance of different solutions, improving the selection efficiency of the target solution.

[0096] In one embodiment, visualizing the Pareto optimal solution set as the Pareto front and presenting it to the production line managers includes:

[0097] Show the optimization objectives to the production line managers and obtain the optimization weights of the optimization objectives set by the production line managers; wherein, the optimization weight is: a numerical value representing the relative importance of the optimization objective;

[0098] According to the Pareto optimal solution set, determine the optimization quality parameter set corresponding to each Pareto optimal solution; wherein, the optimization quality parameter set is: based on the specific numerical values of each quality index after optimization corresponding to the Pareto optimal solution.

[0099] Obtain the initial quality parameter set; wherein, the initial quality parameter set is: the specific numerical values of the current various quality indexes;

[0100] According to the optimization quality parameter set and the initial quality parameter set, determine the quality improvement parameter set; wherein, the quality improvement parameter set includes the improvement degree of each quality index, and the improvement degree of the quality index is specifically: the improvement increment of the initial quality parameter divided by the initial quality parameter value;

[0101] According to the quality improvement parameter set and the optimization weight, determine the optimization value; wherein, the optimization value is: the sum value of the product of the quality improvement parameter corresponding to each quality index and the optimization weight;

[0102] According to the corresponding relationship between the optimization value and the Pareto optimal solution, mark the optimization value correspondingly in the Pareto front.

[0103] The working principle and beneficial effects of the above technical solution are as follows:

[0104] When there are multiple optimization objectives, the optimization preferences of the staff for each optimization objective are different. The production line managers of the present invention can set optimization weights according to the optimization preferences. Determine the set of optimization quality parameters corresponding to each Pareto optimal solution, compare the set of optimization quality parameters with the initial set of quality parameters to obtain quality improvement parameters, and determine the optimization value corresponding to each Pareto optimal solution according to the quality improvement parameters and the set optimization weights; according to the correspondence between the optimization value and the Pareto optimal solution, mark the optimization value corresponding in the Pareto front. The production line managers can intuitively view the optimization value of each Pareto optimal solution quantified according to the self-set optimization weights, which is convenient for the subsequent selection of the target solution and is more user-friendly.

[0105] In one embodiment, based on the intelligent decision result, the corresponding control of the painting production line further includes:

[0106] Obtain the first painting process of the first painting target on the painting production line; wherein, the first painting target is: the painted product on the painting production line; the first painting process is: each time node corresponding to different painting stages (such as: preparation, painting, drying);

[0107] Obtain the switching moment of the intelligent decision result; wherein, the switching moment is: the moment when the painting parameters are optimized and adjusted;

[0108] Determine the second painting target according to the switching moment and the first painting process; wherein, the second painting target is: the corresponding first painting target for optimizing and adjusting the painting parameters during the painting process determined according to the switching moment and the first painting process;

[0109] Extract the image features of the second painting target to obtain the image features; wherein, the image features are extracted according to the captured image of the second painting target, and specifically include: color, texture, shape, etc.;

[0110] Determine the painting effect of the second painting target according to the image features; when determining the painting effect, obtain the evaluation model of the painting stage of the second painting target, and determine the painting effect according to the image features and the evaluation model;

[0111] If the painting effect is less than or equal to the preset painting effect threshold, readjust the second painting target. The preset painting effect threshold is set manually in advance.

[0112] The working principle and beneficial effects of the above technical solution are:

[0113] After the products on the production line are painted, quality inspection is required. Due to problems such as control errors and judgment errors in the optimization timing, when the painting parameters change, the products sprayed by the production line equipment are more likely to have quality problems. Therefore, the present invention obtains the first painting process of the first painting target. In addition, the switching moment of the intelligent decision result is obtained, and the second painting target being painted at the moment when the painting parameter optimization occurs is determined; image features of the second painting target are extracted to obtain the image features; an evaluation model of the painting stage of the second painting target is introduced, and the painting effect is determined according to the image features and the evaluation model; the second painting targets with the painting effect less than or equal to the preset painting effect threshold are screened for readjustment, improving the screening efficiency of the problem-painted products.

[0114] In one embodiment, extracting the image features of the second painting target to obtain the image features includes:

[0115] According to the switching moment and the first painting process, the intervention process point is determined; wherein, the intervention process point is: the moment of painting parameter optimization;

[0116] According to the intervention process point, the intervention painting point of the second painting target is determined; wherein, the intervention painting point is: the painting area point where the second painting target corresponding to the moment of painting parameter optimization is being painted;

[0117] The target section corresponding to the intervention painting point for the second painting target is obtained; wherein, the target section is tangent to the surface of the second painting target, and the tangent point is the intervention painting point;

[0118] The projection vector of the target section is obtained; the straight line where the projection vector is located is perpendicular to the target section, and the vector direction of the projection vector is from the intervention painting point and towards outside the second painting target;

[0119] The shooting vector of the shooting device is obtained; wherein, the vector direction of the shooting vector is: perpendicular to the shooting lens and towards outside the lens;

[0120] If it is satisfied that for any projection vector, there is a vector angle between the shooting vector and the projection vector greater than or equal to the preset angle threshold, and when there is a vector angle between the shooting vector and the projection vector greater than or equal to the preset angle threshold, the corresponding shooting device is triggered to shoot, then the image features are obtained according to the corresponding shooting images.

[0121] The working principle and beneficial effects of the above technical solution are:

[0122] The present invention determines the intervention painting points for the second painting target, and determines a target cutting plane that takes the intervention painting points as the tangent points and is tangent to the surface of the second painting target; introduces the projection vector of the target cutting plane, and at the same time, introduces the shooting vector of the shooting device. If it satisfies that for any projection vector, the vector angle between the shooting vector and the projection vector is greater than or equal to the preset angle threshold, it indicates that there is a suitable shooting situation for each intervention painting point that needs to be shot; when the vector angle between the shooting vector and the projection vector is greater than or equal to the preset angle threshold, the corresponding shooting device is triggered to shoot, indicating that an image is taken every time a suitable shooting situation is met. Image features are obtained based on all the captured images, improving the suitability and comprehensiveness of image feature acquisition.

[0123] In one embodiment, according to the image features, determining the painting effect of the second painting target includes:

[0124] Obtain the target attributes of the second painting target; where the target attributes are: the attribute information of the second painting target, such as: the type of painting object;

[0125] According to the target attributes, obtain the painting evaluation record; where the painting evaluation record is: the evaluation record information of the painting staff evaluating the second painting target;

[0126] Obtain the painting stage of the second painting target; where the painting stage is: different steps in the painting process, such as: primer painting, topcoat painting, drying, curing, etc.;

[0127] According to the painting stage, determine the target evaluation record item in the painting evaluation record; where the target evaluation record item is: the evaluation record information corresponding to the painting stage of the intervention process point;

[0128] According to the target evaluation record item, train the evaluation model; where the evaluation model is obtained by training with the target evaluation record item model;

[0129] According to the intervention painting points, determine the evaluation prompt of the evaluation model; where the evaluation prompt is: the position guidance information when the evaluation model makes an evaluation;

[0130] After the evaluation prompt is completed, input the image features into the evaluation model to determine the painting effect of the second painting target.

[0131] The working principle and beneficial effects of the above technical solution are:

[0132] The present invention obtains the painting evaluation record of the painting staff evaluating the second painting target according to the target attributes of the obtained second painting target; determines the target evaluation record item in the painting evaluation record according to the painting stage of the second painting target determined by the painting task process; trains an evaluation model applicable to the painting stage of the second painting target based on the target evaluation record item, which is more suitable; since the spray painting effect evaluation is greatly affected by the shape of the second painting target, according to the intervened painting point positions, the evaluation prompts of the evaluation model are determined to guide the evaluation positions of the painting evaluation by the evaluation model. After the evaluation prompts are completed, the image features are input into the evaluation model to determine the painting effect of the second painting target, and the painting effect evaluation is more refined.

[0133] The painting production line provided by the embodiment of the present invention is applied to the method described in the above embodiment.

[0134] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalent technologies, the present invention also intends to include these changes and modifications.

Claims

1. Intelligent decision-making method for coating production line, characterized in that: include: Determine the useful life period of the painted parts on the painting line; Filter target painting records according to the service life period, and train intelligent decision-making models based on the target painting records; Perform multi-objective optimization based on intelligent decision-making models to obtain intelligent decision-making results; Acquire a first coating process of a first coating target on a coating production line; The switching moment for obtaining intelligent decision-making results; Determine the second coating target according to the switching time and the first coating progress; Determine the intervention process point according to the switching time and the first coating process; According to the intervention process point, determine the intervention painting point of the second painting target; Acquire a target section corresponding to the second coating target at the intervening coating point; Obtaining a projection vector of the target section; the straight line where the projection vector is located is perpendicular to the target section, and the vector direction of the projection vector is from the intervening coating point to the outside of the second coating target; Obtaining a shooting vector of a shooting device; If it is satisfied that for any projection vector, the vector angle between the shooting vector and the projection vector is greater than or equal to the preset angle threshold, and the corresponding shooting devices are triggered to shoot when the vector angle between the shooting vector and the projection vector is greater than or equal to the preset angle threshold, then image features are obtained according to the corresponding shot images; Get the target attributes of the second painting target; Obtain coating evaluation records based on target attributes; Get the painting stage of the second painting target; Determine the target evaluation record items in the coating evaluation record according to the coating stage; According to the target evaluation record items, the evaluation model is trained; Determine the evaluation prompts of the evaluation model according to the intervention coating points; After the evaluation prompt is completed, the image features are input into the evaluation model to determine the coating effect of the second coating target; If the painting effect is less than or equal to a preset painting effect threshold, the second painting target is readjusted.

2. The coating production line intelligent decision-making method according to claim 1, characterized in that: Determine the useful life stage of the painted parts on the painting line, including: Obtain equipment maintenance records for coating production lines; Obtain the usage time of the painted parts based on the equipment maintenance records; Get the consumable type of the painted part; Determine the rules for the life cycle of the painted parts based on the type of consumables; Determine the rules and usage duration based on the life cycle and determine the usage life period.

3. The coating production line intelligent decision-making method according to claim 1, characterized in that: According to the service life period, the target painting records are screened and the intelligent decision model is trained based on the target painting records, including: Obtain pre-selected painting records; pre-selected painting records include: reference painting records obtained by big data and historical painting records on the painting production line; Parse the pre-selected painting records to determine the reference service life period of the reference painted parts; Determine the target coating record based on the service life period and the reference service life period; Parse the target painting record to obtain a training data set; the training data set includes: a one-to-one corresponding training painting parameter set and a training painting quality parameter set; Based on the preset CNN model, the intelligent decision-making model is trained according to the training data set.

4. The coating production line intelligent decision-making method according to claim 1, characterized in that: Based on the intelligent decision-making model, multi-objective optimization is performed to obtain intelligent decision-making results, including: Based on the intelligent decision-making model, the optimization algorithm is used to perform multi-objective optimization to obtain intelligent decision-making results.

5. The coating production line intelligent decision-making method according to claim 4, characterized in that: Based on the intelligent decision-making model, the optimization algorithm is used to perform multi-objective optimization to obtain intelligent decision-making results, including: Get optimization target; Based on the intelligent decision-making model, the Pareto optimal solution set is determined according to the optimization algorithm and optimization goal; The Pareto optimal solution set is visualized as a Pareto frontier and presented to the production line managers, and intelligent decision-making results are obtained based on the target solution selected by the production line managers.

6. The coating production line intelligent decision-making method according to claim 5, characterized in that: Visualize the set of Pareto optimal solutions as a Pareto front and present it to production line managers, including: The optimization target is presented to the production line manager, and the optimization weight of the optimization target set by the production line manager is obtained; According to the Pareto optimal solution set, determine the optimization quality parameter set corresponding to each Pareto optimal solution; Obtain an initial quality parameter set; Determine a quality improvement parameter set according to the optimized quality parameter set and the initial quality parameter set; Determine the optimization value based on the quality improvement parameter set and optimization weight; According to the corresponding relationship between the optimization value and the Pareto optimal solution, the optimization value is marked accordingly in the Pareto frontier.

7. The coating production line intelligent decision-making method according to claim 1, characterized in that: Based on the intelligent decision-making results, the coating production line is controlled accordingly, including: Get the current painting parameter set; Determine the control factors of the coating production line based on the current coating parameter set, intelligent decision-making results and the preset control factor library; Control the coating production line for production based on the control factors.

8. Painting production line, characterized in that, The method according to claims 1 to 7 is applied.

Images